createProbMatrix: Probabilities and joint probabilities of assignment.

In ck37/htestimate: Horvitz-Thompson Estimate

Description

Return a matrix of probabilities and joint probabilities of assignment for n observations and k randomization replications.

Usage

createProbMatrix(assignments, byrow = F)

Arguments

assignments	Each column is a replication and each element is the unit's assignment.
byrow	Change the assignment matrix to use rows rather than columns for each randomization.

Value

n*k by n*k matrix of probabilities and joint probabilities of assignment to each arm.

Examples

# Create test matrix 1 per the PDF document.
arms1 = 1:3
# We transpose the results so that each permutation is a column.
testmat1 = t(crank::permute(arms1))
# Display the simple test matrix.
testmat1
prob_matrix = createProbMatrix(testmat1)

# RI package example, but without blocking or clustering.
y = c(8,6,2,0,3,1,1,1,2,2,0,1,0,2,2,4,1,1)
z = c(1,1,0,0,1,1,0,0,1,1,1,1,0,0,1,1,0,0)
# Create 10k random permutations of the assignment vector.
perms = ri::genperms(z, maxiter=10000)
prob_matrix = createProbMatrix(perms)

ck37/htestimate documentation built on May 13, 2019, 7:34 p.m.